TECHNICAL FOCUS: PRODUCT IN DEPTH

TECHNICAL FOCUS:

PRODUCT IN DEPTH

We have definitely reached a plateau in automated LED pro-file spots in that there is now a tacitly agreed-upon outline specification that all manufacturers are working to, with dif-ferences in their own particular special features. That broad specification is for a unit with a white LED engine, CMY + CTO color mixing, one or two color wheels, one or two gobo wheels, framing, and iris, plus beam effects. What we’ve seen is the brightness of white LED engines rise year-on-year as the technology improves, and it hasn’t stopped going up yet. We are seeing an arms race, with lumens as the target. The downside—to me, at least, as I get older and struggle to lift them out of their road cases: While the first LED-based fixtures were lightweight compared to their HID lamp predecessors, they are now back to similar weights. They use less power, produce less heat, and don’t need new lamps every 500 hours, but the weight has crept up. As LEDs have become the norm, product differentiation is now in the details of the engineering, special effects offered, and

innovations to help make the user’s life easier.

This month, I’m looking at the new Robe FORTE LED luminaire. As with the last Robe product we covered, the Esprite, the Forte uses the company’s in-house LED white light engine in which the LEDs are separate from the homogenizing and collimating optics, allowing placement of the color-mixing wheels in an optimal place, in the middle of those optics, for smooth mixing. This system also provides the ability to remove and replace the LED engine relatively easily. I think the jury is still out as to whether this will be a compelling feature for Robe, but I do appreciate the improved color mixing. It probably depends on your place in the supply chain. The lighting designer will look at the color mixing and will like units that match in brightness but may not care about the removable engine. A rental company might look at serviceability, the ability to switch out different color-rendering engines, and the ability to match light engines when making a buying decision. As to the weight, the Forte is 88lb, definitely lighter than others.

The results presented here are based on my tests of a Forte supplied to me by Robe North America (Figure 1). I ran the fixture on a nominal 120V 60Hz supply, but the unit is self-adjusting for supply voltage and will run on any volt-age from 100-240V, 50/60Hz.

Light source

Figure 2 shows the 1,000W HP white light LED light engine after I removed it from the fixture. It’s very similar in layout to the Esprite, except that its engine had 73 LEDs while this one has 121. The LEDs and their circuit board—as well as the copper heat spreader, heat pipes, and heat sink—come out as a single unit after removing eight screws and unplug-ging three connectors. It took less than five minutes. I was also able to download the Robe app to my phone and inter-rogate the NFC system on the module. This is done without any power being required. Figure 3 shows the result, the data on that specific light engine, including its history and run time. The Forte also has a light level sensor mounted just inside the final output lens housing. Robe tells me that

Robe FORTE LED Luminaire

Figure 1: Fixture as tested.

Figure 2: LED engine.

stores the data in both the light engine and the main Forte electronics.

Figure 4 shows the first lens array where the LED engine is attached. These are collimating lenses, one per LED, that direct the light through the color-mixing system before the beams are homogenized. When the LED engine is back in place, the heat sink aligns with the six cooling fans mounted to the fix-ture, as shown in Figure 5.

Just after those

collimating lenses, on the other side of the bulkhead are the color-mixing dichroics; more on them later. They are fol-lowed by two sets of fly-eye lenses, arranged in hexagonal arrays that provide beam homogenization, and a final con-denser lens that directs the light through the remainder of the optical system. You can see the condenser and fly-eye lenses behind the two color wheels in Figure 6. Note that, unlike the collimating lenses, there isn’t a one-to-one rela-tionship between the fly-eye lenses and the LEDs. There are many more of them than LEDs so that distribution overlaps more than one lens. Effectively, each fly-eye lens samples the output of multiple emitters and mixes them together. Color systems

As already mentioned, the Forte uses a dichroic mixing system mounted between the two sets of LED lenses. This arrangement helps improve the uniformity and smoothness of the color mixing. There are four pairs of flags in cyan, magenta, yellow, and CTO. The transmis-sion of each color is shown in the table below. These are fairly saturated colors.

COLOR MIXING

Color Cyan Magenta Yellow Red Green Blue CTO

Output 17% 6.5% 74% 6.0% 9.2% 0.2% 42%

As with the Robe Esprite and other products in the Robin range, the color mixing can be controlled via a virtual color

wheel channel as well as through normal CMY controls. The color mixing is very smooth, and I saw very little colored edging on defocused gobos with pastel colors.

Next in line are two color wheels, concentrically mounted on a single shaft, as shown in Figure 6. As you can see from the photograph, each color filter is shaped as a skewed trapezoid. Robe has done this to compensate for the center of the wheels being offset from the centerline of the

lumi-Figure 3: NFC data from light engine.

Figure 4: Collimating lenses.

Figure 5: Cooling.

TECHNICAL FOCUS:

PRODUCT IN DEPTH

naire so that the line between half colors will be vertical through the center of the beam, as shown in Figure 7.

COLOR WHEEL 1

Color Deep Red Deep Blue Orange Green Congo

Output 0.8% 0.6% 13% 15% 0.1%

COLOR WHEEL 2

Color Multicolor Lime Green Lavender CRI 80 CRI 90

Output N/A 71% 24% 77% 67%

I measured the spectrum and color rendering of the Forte in four situations: in open white, with full CTO, and with both CRI filters. These results are shown in Figures 8, 9, 10, and 11 and are tabulated below.

Filter CCT TM-30 Rf TM-30 Rg CRI

Open White 6972K 67 95 69

Full CTO 2976K 71 97 71

CRI 80 7330K 75 102 81

CRI 90 6568K 82 103 86

As we often see with white LEDs, the Forte oversaturates colors slightly in green and magenta.

Note: The unit I tested was fitted with the HP, or high-performance, LED engine. Robe also has an alternative engine called HCR, for high color rendition. I was not able to test it, but Robe claims it has a CRI of 94 and about 60% of the light output of the HP engine.

The remainder of the Forte’s optical effects are mounted on two removable modules, one for the gobos and anima-tion wheel, and one for the framing shutters and iris. Gobo module

Figures 12 and 13 show the two sides of the gobo module. The two rotating gobo wheels are identical, each having six replaceable glass gobos supported in removable snap in holders.

ROTATING GOBO SPEEDS

Gobo change speed – adjacent 0.5 sec Gobo change speed – worst case 0.8 sec

Maximum gobo spin speed 0.38 sec/rev = 159 rpm Minimum gobo spin speed 1784 sec/rev = 0.03 rpm Maximum wheel spin speed 3.6 sec/rev = 17 rpm Minimum wheel spin speed 54 sec/rev = 1.1 rpm

Rotation and indexing on both wheels were very smooth. I measured the repositioning accuracy at 0.05° of hysteresis error, which equates to 0.2" at a throw of 20'. Figure 14 shows an example of a gobo morph from one gobo wheel to the other.

Figure 7: Half colors.

Last in this module is the animation wheel. There’s noth-ing unusual here; this is the standard Robe breakup anima-tion wheel. I measured it taking 0.4 seconds to insert or remove the wheel and, once in place, it can be rotated at speeds varying up to 1.8 sec/rev = 33rpm.

Framing module

Figures 15 and 16 show the two sides of the framing shutter module and the iris. Each shutter blade has approximately +/- 21º of rotation and can move in to cover about 75% of the beam, taking about 0.3 seconds to move in or out.

Figure 13: Gobo module rear.

Figure 14: Gobo morph.

TECHNICAL FOCUS:

PRODUCT IN DEPTH

Figure 17 shows the range of shutter angle rotation. The entire shutter system can also be rotated +/- 60º. The differ-ence between focus on the four blades can be seen in Figure 18. In this photograph, I hard-focused on the right-side blade. This figure also shows good straight edges to the framing with relatively little pincushion or barrel distor-tion. There is some at extreme zoom ranges as you would expect, but nothing unusual.

Last in this module is the iris. This can open or close in 0.25 seconds and reduces the aperture to 17.5% of its full size. This results in equivalent field angles of 0.9° at mini-mum zoom and 8.5° at maximini-mum zoom.

Prism and frost

The Forte has two indexable, rotatable prisms mounted to the output side of the rear lens group. These travel back and forwards with that lens. Both prisms can be inserted separately or simultaneously, and their movement synchro-nized. Figure 19 shows the prisms and Figure 20 shows an example of the kind of effect that can be achieved by over-laying the two prisms—in this case, with the multicolor filter and a gobo. Because the lenses may have to move, insert-ing either prism can take up to 1.5 seconds. They can then be rotated at speeds up to 115rpm or down as low as

0.13rpm in either direction.

Finally, we have two frost flags. These are mounted to the input side of the final lens group, just after the prisms as shown in Figure 21. Figures 22 and 23 show the effect on a gobo at 0%, 50%, and 100% for each of the two flags. Both flags are changeable through a magnetic attachment and can be inserted or removed in 0.4 seconds. Both frost flags are on the same DMX channel so you can’t insert both flags at the same time.

Dimming and strobe

The dimming on the Forte is very smooth and clean, even at the bottom end. Figure 24 shows an example of the linear curve. PWM defaults to 600Hz but optional frequencies, up

Figure 17: Shutter angle range.

Figure 18: Shutter focus.

Figure 19: Prisms.

September and 1,500Hz from 2023.) Strobe range is

adjustable from 0.5Hz up to a measured 18Hz.)

Lenses and output

Output measurements were taken with the fans in auto mode after the unit had reached thermal equilibrium. I measured the thermal droop of the Forte at a maximum of 6% after running for five minutes at full power. This is a very good result and suggests that Robe is doing some thermal calibration. After that 6% drop, there were no further changes in output.

Figure 21: Frost flags.

Figure 24: Dimmer curve.

Figure 22: Frost 1.

TECHNICAL FOCUS:

PRODUCT IN DEPTH

The Forte uses the ubiquitous three-group lens system: two groups that move for zoom and focus, and a fixed out-put group. I measured the zoom group as taking 0.8 sec-onds to travel from end to end, with the focus group taking 0.6 seconds. The light output in wide angle (after reaching thermal equilibrium) was just under 34,000 lumens at a field angle of 48.5°. At narrow angle, I measured 18,100 lumens at 5.2° field angle. Figures 25 and 26 show the beam distri-bution which is extremely flat.

Note: I measure light output using a flat-wall method and measure the throw from the hanging point. My results will differ from those from a goniometer. Firstly, I always keep my meter flat to the wall rather than pointing it at the light source, as happens with a goniometer; secondly, with a flat-wall method, the throw distance increases at the edges of

the beam. This makes a significant difference at very wide angles. I do have and use a goniometer as well, but I believe that a flat-wall measurement more closely replicates what a user sees on a stage with these kinds of luminaires. A goniometer will make wide-angled beams look flatter than I show. I’ve always used this method for reviews in this maga-zine and, even though my data may not agree with the man-ufacturers, it is internally consistent and allows comparing one of my reviews with another.

Pan and tilt

I measured the pan and tilt range of the Forte at 540° and 270°, respectively. A full-range 540° pan move took 4.4 sec-onds to complete, while a more typical 180° move finished in 2.5 seconds. Tilt took three seconds for a full 270° move and 2.4 seconds for 180°. Movement in both axes is very precise with no visible ringing or wobble on stopping. I measured hysteresis on both pan and tilt at 0.05°, equiva-lent to 0.2" at 20'.

Noise

With 1,000W of LEDs to keep cool, the fans provide the noise floor for the Forte. Pan and tilt, along with zoom and focus, were the noisiest movements. However, the loudest overall was moving a prism into place, as this can also include moving both lenses to make space.

SOUND LEVELS Normal Mode Ambient <35 dBA at 1m Stationary 48.7 dBA at 1m Homing/Initialization 52.9 dBA at 1m Pan 52.2 dBA at 1m Tilt 51.8 dBA at 1m Gobo 48.9 dBA at 1m Zoom 50.4 dBA at 1m Focus 51.6 dBA at 1m

Animation wheel 48.9 dBA at 1m

Framing 51.9 dBA at 1m

Frost 48.9 dBA at 1m

Prism 55.2 dBA at 1m

This was with the fans running in auto mode. Fan high-speed mode increased the noise level considerably but with very little increase in output. Setting the fans to their lowest speed made a big difference to noise, dropping it by 6dB, while output dropped to 88%.

Homing/initialization time

Full initialization took 91 seconds from a cold start and 69 seconds from a DMX-512 reset command. Homing is well-behaved in that the fixture fades out smoothly, resets, and keeps the LEDs off before fading up again after all reset movement is finished.

Figure 25: Maximum zoom.

removable covers, mod-ules, and distributed motor drive boards. Removing the two main optical mod-ules was simple. I would expect basic maintenance and cleaning to be straightforward. Figure 27 shows the overall head layout.

Figure 28 shows the two yoke arms with the covers removed revealing the pan and tilt motors and tilt belt system.

Electronics and control

The Forte uses the standard Robe color touch-screen sys-tem, which provides access to a comprehensive array of setup and service functions (Figure 29). This includes RDM (tested using City Theatrical DMXcat), Ethernet protocols, optional wireless DMX using the LumenRadio CRMX sys-tem, stand-alone operation, and self-test modes.

The connector panel on the opposite side of the top box contains Neutrik True-1 TOP (the black outdoor version that

replaced the now obsolete standard True-1) power input along with standard five-pin and three-pin DMX512 connec-tions and an Ethernet port (Figure 30). Robe tells me that, as an option, a user can have the Forte equipped with an Ethernet out port. In that case, there is new circuitry in the unit that ensures the Ethernet output passes data even if the fixture loses power.

I measured power consumption when running at full out-put open white as 10.4A, 1186W, 1207VA, a power factor of 0.99. The quiescent load with all LEDs off was 0.87A, 105W, 107VA, power factor of 0.97.

We’ve gone from input to output on the Robe Robin Forte. It showcases the next version of their in-house removable LED light engines with many of Robe’s standard features. Is the Forte for you? I encourage you to get a demonstration if it looks interesting. The final word is always yours.

Mike Wood provides design, research and development, technical, and intellectual property consulting services to the entertainment technology industry. He can be contacted at [email protected].

Figure 27: Head.

Figure 28: Yoke arms.

Figure 29: Display.